Project Summary The goal of this project is to develop computational pipelines that allow in silico prediction of functional genetic variants that disrupt pre-mRNA splicing and related pathways in Alzheimer's disease (AD). Recently, tremendous success has been achieved in constructing a catalog of genetic variants in AD genomes of various patient cohorts. The next great challenge is to identify causal variants and elucidate their potential function relevant to disease processes. To this end, research efforts have been directed to studying variants located in protein-coding, promoter, and splice site regions due to their apparent impacts on gene expression. However, many of the newly identified disease-associated variants reside in other non-coding regions, such as introns, that may confer regulatory function to the related gene. The mechanisms of these variants have been hard to decipher. It is expected that many of them may function at the post-transcriptional level, thus affecting mRNA expression. In human, a myriad of processes mediate RNA expression at the post-transcriptional stage, one of which being splicing. Splicing is an essential step of mammalian gene expression and alternative splicing affects most human genes. Recent literature reported that RNA splicing is a primary link between GVs and disease. In general, it was estimated that 15-60% of point mutations that result in human genetic diseases disrupt splicing, highlighting the importance of this regulatory step. In AD, aberrant splicing has been detected in many functionally critical genes, some of which are modulated by GVs. Despite the importance, how to accurately identify functional genetic variants in splicing regulation remains a key question in the field. To address this question, the large collection of RNA-Seq and genotyping data sets collected from AD and control subjects represent an invaluable resource. We will develop and apply novel methodologies to make full use of these data sets, complemented by further bioinformatic prediction and experimental validations. This work will allow a previously unattained level of understanding of genetic variants in splicing regulation and provide new means to tackle the imperative task of functional annotations of genetic variants in AD.